Sound/Sign Patterning (Phonology)

The sound segments (phones) of a spoken language are grouped into categories known as phonemes. A phoneme is a category of sounds that are treated as representationally equivalent in a given language system, even though their physical realization may differ. For example, the following sounds in American English are treated equivalently as the phoneme /t/ (phonemes are conventionally written between slashes, and phones in square brackets), rendered as "t" in English orthography: top, stop, pot, potter, button. In many dialects of American English, these five sounds are all articulated differently, and their acoustic properties differ as well, yet they are all treated as valid exemplars (allophones) of the same phoneme /t/. The type of variant (allophone) that occurs in a given environment (e.g., beginning of a word, after an [s], end of a word, between vowels, before an [n]) is usually not random but predictable. These so-called alternations can be captured by phonological rules that specify which variant (allophone) of a phoneme will surface in a given environment. Common phonological rules also specify when phones are inserted or deleted in certain environments, when they become phonetically more or less like adjacent phones, when they reduplicate or switch places with other phones, how they are grouped together to form syllables, etc. These processes constitute "segmental phonology," which deals with the behavior of individual speech segments (cf. "Segmentation"). At higher ("suprasegmental") levels of analysis, phonological processes also describe the predictable behavior of stress (accent), pitch, tone (in tone languages), etc.  No naturally occurring system of animal vocalization has thus far demonstrated this high degree of elaborated internal hierarchical structure (cf. "Hierarchical structure") or these types of intricate rule-governed behaviors. Since the sounds of human language are dependent on and arise from the structure of the human vocal tract, animal vocal tracts are not capable of producing the sounds of human language to this degree of specificity, and therefore cannot be trained to reproduce it accurately. Since sign language neither depends on nor arises from the configuration of the human vocal tract, it does not exhibit exact parallels to spoken phones and phonemes. However, it is similarly organized around segments (handshape, hand orientation, movement in space, and body location), which as in spoken languages can be inserted or deleted in certain environments. Combinations of these segments occurring over time have also been used to define the sign equivalent of the syllable. In addition to sharing many of the same types of phonological processes and rules as spoken language, sign languages also manifest a suprasegemental level of analysis (often tied to linguistic markers of facial expression that accompany manual signing).